Crystal Chang Din

My research interests lie in the areas of formal methods, software verification, and didactics.

I am a member of ECOOP 2021 Artifact Evaluation Committee, UDIT 2021 Program Committee and TheWebConf 2022 Program Committee.

Available master thesis topics:

In software engineering, demand of customers for configuration options that address various different business concerns creates the need to manage variability by developing not just a single software system but, in fact, an entire family of software systems with similar functionality. Software Product Line (SPL) engineering is a software engineering method to efficiently develop a family of software systems by capitalizing on their similarities while explicitly handling their differences. Due to the size and complexity of these systems, SPLs constitute a major investment with long-term strategic value. Over time, SPLs have to be adapted as part of software evolution to address new requirements, which is particularly complicated as an entire software family has to be adapted. 

While existing approaches (https://dl.acm.org/doi/10.1145/3382025.3414964) can identify the evolution paradoxes in an evolution plan of SPLs and can analyse whether an evolution plan will satisfy the given requirements, there is still lacking support for explaining why the business requirements cannot be satisfied (Thesis A). Furthermore, instead of rigorously prohibiting problematic changes, we may devise a flexible method that determines appropriate additional changes to perform to still reach the intended evolution goal when possible (Thesis B). An SPL can be defined in terms of multiple configurable features structured within a feature model, which also specifies feature dependencies. We would like to extend the current work so that feature dependencies are also taken into account in the reachability analysis of software evolution planning (Thesis C).